I don't know that I disagree with you that much, but nukes present a number of operational difficulties.
For one thing, they're difficult to produce - you need a source of weapons-grade uranium for example, not something that al-Qaeda would be likely to be able purify themselves. So that leaves buying one - certainly they'd have the money, but you still have to find a willing seller ... one whom you can be sure will sell you the real thing and not a mock-up.
Then you have to consider that they're really a pretty fine piece of machinery, with various maintenance and shelf life requirements; and the parts won't be available at the local hardware store.
Then you have to transport them to their target. Certainly you can try to sneak one close to a city on board a ship, but I suspect that we've got a pretty good idea what ships might be compromised. And if your target isn't a seaport (as many big cities aren't), you have to move it around on land. All of this poses risks of discovery since the bomb will be either heavy, radioactive, or both. It's not that it's impossible to escape detection, but it does pose potential risks to the mission.
Now consinder that these are probably single-stage bombs - uranium only, no hydrogen. The hydrogen would introduce additional variables including reduced dshelf life and increased weight, as well as being less likely to be available in a portable bomb. The effect is that they're similar to the Hiroshima and Nagasaki bombs in terms of yield, which means that it's hard to get more than about 100,000 casualties from them except in the very densest cities - and in most places less. So if you really want to cause1 massive destruction, you need more than two or three or even a dozen. >P?The point isn't that any of this is impossible, simply that it introduces unnecessary risks to the mission. Infectious biologicals like smallpox don't have these drawbacks - they're easy to hide before you've released them, and once released they spread on their own -- unlike airborne anthrax for example which will only infect those in the immediate neighborhood of the release.
Look at the things throughout history that have caused the most horrific casualties - the ones at the top of the list are almost always epidemics. Even today we can treat 1only 1the major bacterial infections - many bacterial and essentially all viral diseases are untreatable. The primary defenses againnst them have been containment and vaccination. For some of them we still don't have an effective vaccine, and for others (eg, smallpox) a vaccine exists but relatively few Americans are current on it. Now also consider that they would have no compunction about testing their strain for potency either by a volunteer or a forced subject. In the West this would be considered highly immoral but they don't operate by the same moral code. Finally remember that the release of an infectious agent is much easier than a noninfectious agent like anthrax or a chemical agent.
The implication is clear that you can probably do the most damage with a biological agent, especially if you don't care if the epidemic spreads back to your own country. Maybe you hope to be able to vaccinate people before it gets there or maybe you just don't care.
Finally note that the cost is much lower for biologicals.
My conclusion is that for someone who wants to do the most damage a biological agent is almost irresistable. I'm not sure that this gives me much comfort however. :-(